History repeating itself: Out of Browser

I’ve been following the concept of what RIA represents since well, I can’t honestly remember but it’s around 1998. My first glimpse of RIA was via Erik’s ye olde project called WebOS.com, which was a virtual web based operating system built with DHTML (pre-AJAX wave).

It began with Formats

For years I’ve studied the interactive industry and the twists and turns it has taken. I can even remember the days when SVG came onto the scene and how it was supposed to be the next best thing beyond what VML could offer (not to mention challenged VRML head-on)

It had somewhat of stagnant result, I think Flash offered more interactive and controlled state of play when it came to vector art + animation online in the early days. The secret was tooling though, formats on their own were still considered “amazing science projects”, tooling is what set them free.

As we moved forward in the industry, RIA began to emerge, it was somewhat of a hard pill to make others swallow as there were a lot of product misalignments in those days by Macromedia – I remember sitting in a conference in Port Stephens beating Macromedia staffers up on bad marketing and poor product SKU pricings.

HTML was a strong contender, as I remember sitting in CIO offices asking for $15k+ to replace JavaScript & HTML whilst also explaining that training was required. It was a tough sell.

RIA was born, but became hijacked

AJAX also emerged as being a strong competitor to Flash, so much so that when it’s acronym was first blessed a lot of us die-hard Flash zealots dismissed JavaScript as being this old redundant language that had little value. We were wrong of course, as AJAX not only dominated but essentially hijacked the RIA concept as when you have 90% of the world’s web developers locked into JavaScript as a “I know this language” it’s somewhat hard to fight against that majority ruling (i tried in both camps – Adobe and Microsoft’s and came out bruised).

Before a lot of this, No Touch Deployment (SmartClient) (2002) was a piece of goodness that Microsoft pitched in the early days as well, the basic concept was you could write WinForm like applications that could be deployed via a URL, whereby you open up the .exe via a webrowser and suddenly you had a rich desktop deployed from the web.

SmartClient had potential, but timing was just bad

The downfall for SmartClient I think was essentially the .exe extension and virus/spyware becoming more scary via mainstream press / technical experts. Imagine this, you load a URL, and a .exe is presented to you, you then agree to install it and it now has access to your computer within a specialized sandbox.

Not to mention Java Webstart was also a strong contender for the Java camp, given it’s seductive footprint around allowing SWING clients to launch via the web.

Later in 2003, Macromedia Central beta was released, which at first glance had a lot of promise. It was simply a Chrome framework in which you socket your Flash .SWF into and then it becomes a widget style application you can run, housed safely within a specialized sandbox.

It was much like the AppStore model you see today in the iPhone as well. Macromedia were onto something there, but the reality was that developers appeared to want to take it beyond the browser, and so Macromedia/Adobe camp went silent on the product for a while.

Out of Browser is an old concept in new technology

Fast forward to the present, Adobe AIR hails from the central DNA and Microsoft has put forward to options that hail from SmartClient – WPF and Silverlight.

WPF we have two flavors, .XBF (very similar to SmartClient strategy but more focused in safety and XAML) and of course the full blown .exe native desktop footprint.

The problem however, is that with reach comes a need, a need to be treated as a serious application but housed snuggly within a safe sandbox (i.e. see reasons for SmartClient failure to launch as to why).

Silverlight Out of Browser didn’t hail from Adobe AIR, and whilst it’s easy to look at the two options today at face value and go “ahah!..see its a copy..” the reality is this problem has been iterated on since 2002 and below 2002.

I think patterns are re-emerging from the past, and you’ll see us all taking old ideas and improving them as technology tree begins to open up new branches. Whilst one brand may have success in proving the concept has market potential today, it doesn’t necessarily mean that their competitors are just waiting and seeing. Innovation happens daily in both brands, and to simply associate a concept with one single brand is very limiting to what both had to offer the world in the past.

History repeating itself

In 1998 I studied VRML (Virtual Reality Markup Language) and used to code this beast of a language on a 30k O2 machine (given at the time GPU was highly expensive). The concept was simple, 3D web… it failed as the technology tree wasn’t ready for it.

Today, you see 3D (fake and real) in not only plug-in’s but now desktop clients. Sun’s Silicon Studios has a lot to answer for, as by crikey they had it wide open for pickings..

In 1998 I remember having Novell network on Windows 3.11 whereby you would have applications made available to you remotely via a network. Today we have Windows Azure pushing the same concept but more highly developed.


This industry is constantly re-innovating old ideas into new ways or approaches. I think the waters of purity around whom owned what and where are so murky that you’ll be hard pressed to argue how cleanly one is better than the other.

I’ve spoken to many staff from all brands (Apple, Adobe, Microsoft, Macromedia, Sun, Google etc) about this very conversation and it’s amazing to hear how certain products derive inspiration from one another.

As Pablo Picasso once said:

Bad artists copy. Good artists steal

Scott Out.